Human breast cancer cell lines BT-474 and MDA-MB-231 and the embryonic kidney cell line 293T were purchased from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in Dulbecco’s modified Eagle’s medium (Life Technologies; Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% fetal bovine serum (Life Technologies; Thermo Fisher Scientific, Inc.) in a humidified atmosphere of 5% CO₂ at 37°C. Paclitaxel was purchased from Selleck Chemicals (Houston, TX, USA). Polybrene and puromycin were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany).

Cells (1.0×10⁴ cells/well BT-474 cells or 3.0×10³ MDA-MB-231 cells) were seeded into 96-well plates. After 24 h, the cells were exposed to different concentrations of paclitaxel (0.001, 0.01, 0.1. 0.5, 1, 5, 10 and 20 µM for BT-474 cells, or 0.001, 0.002, 0.005, 0.01 and 0.02 µM for MDA-MB-231 cells) for 72 h at 37°C, unless otherwise stated. Each concentration was repeated in quadruplicate wells. Subsequently, cell viability was measured by adding 10 µl CCK-8 reagent (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) to each well. Plates were incubated at 37°C for 1 h and absorbance was evaluated using a Bio-Rad microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 450 nm wavelength. The percentage of viable untreated control cells was set as 100%.

Subsequent to 0.5 or 0.002 µM paclitaxel treatment for 72 h at 37°C for BT-474 or MDA-MB-231 cells, respectively, they were collected and stained using the Annexin V-fluorescein isothiocyanate (FITC) kit (Miltenyi Biotec, Inc., Cambridge, MA, USA) according to the manufacturer’s protocol. Briefly, cells were resuspended in binding buffer (100 µl) and stained with Annexin V-FITC (10 µl) for 15 min at room temperature in the dark. Subsequent to adding propidium iodide solution (5 µl), the apoptosis rate was immediately determined by calculating the percentage of cells that were positive for Annexin V, as measured using the FACS Calibur flow cytometer and CellQuest™ software (version 6.0; BD Biosciences, Franklin Lakes, NJ, USA). All experiments were performed three times.

The BT-474 and MDA-MB-231 cell lysates were subjected to western blot analysis. Anti-Beclin1 (cat. no. sc-11427; 1:2,000 dilution) antibody was obtained from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Anti-caspase-3 (cat. no. 9665; 1:1,000 dilution), anti-cleaved caspase-3 (cat. no. 9664; 1:1,000 dilution), anti-poly ADP ribose polymerase (PARP; cat. no. 9542; 1:1,000 dilution), anti-cleaved PARP (cat. no. 5625; 1:1,000 dilution) and anti-Actin (cat. no. 4967; 1:1,000 dilution) antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Western blot analysis was performed as previously described (32). Actin was used as a protein loading control. The optical density of each protein band was quantified using ImageJ software version 1.52 (National Institutes of Health, Bethesda, MD, USA).

Total RNA extracted from BT-474 and MDA-MB-231 cells was isolated using TRIzol (Life Technologies; Thermo Fisher Scientific, Inc.) and 2 µg total RNA was reverse transcribed using Superscript™ III transcriptase (Life Technologies; Thermo Fisher Scientific, Inc.). For RT-qPCR, cDNA was amplified using THUNDERBIRD SYBR qPCR Mix (Toyobo Life Science, Osaka, Japan) in a Chromo4™ system (Bio-Rad Laboratories, Inc.). PCR was initiated with a 10-min denaturation step at 95°C, followed by a 40-step cycle at 95°C for 15 sec and 60°C for 60 sec. The primers used were as follows: Beclin1 forward, 5′-TCTCGCAGATTCATCCCC C-3′ and reverse, 5′-TCTTCGGCTGAGGTTCTCCAT-3′; GAPDH forward, 5′-ACGGATTTGGTCGTATTGGG-3′ and reverse, 5′-TGATTTTGGAGGGATCTCGC-3′. The 2^−ΔΔCq method was used to calculate the relative gene expression (33). GAPDH was used as an internal control for normalization.

An RNA interference technique was used to knockdown Beclin1 in breast cancer cells, as previously described (34,35). The backbone vector used for generating the lentivirus was pGC-LV, purchased from Shanghai GeneChem Co., Ltd. (Shanghai, China). The sense strands of the short hairpin RNAs (shRNAs) were as follows: shControl, 5′-TTCTCCGAACGTGTCACGT-3′; shBeclin1#1, 5′-CAGTTTGGCACAATCAATA-3′; and shBeclin1#2, 5′-CCCGTGGAATGGAATGAG ATT-3′.

To produce infectious viruses, 3×10⁶ 293T cells were seeded on 100-mm plates and the day after cotrans-fected with lentiviral backbone plasmid and packaging plasmids (pHelper 1.0 and pHelper 2.0). All transfections were performed using Lipofectamine^® 3000 (Thermo Fisher Scientific, Inc.) according to the manufacturer’s protocol. The vector pGC-LV was used as an empty vector control to assess transfection efficiency. After 48 h transfection, the medium was collected, purified, aliquoted and stored at -80°C. Virus titer was determined with a Lenti-X™ p24 Rapid Titer kit (cat. no. 632200; Clontech Laboratories, Inc., Mountain View, CA, USA) according to the manufacturer’s protocols. For infections, BT-474 and MDA-MB-231 breast cancer cells plated in 60-mm plates for 24 h prior to infection were incubated with lentivirus at a multiplicity of infection of 20 in the presence of polybrene (8 µg/ml) for 48 h at 37°C. After, the media containing 1 µg/ml puromycin was added to select stably infected cell populations.

BT-474 (4,000 cells/well) or MDA-MB-231 (500 cells/well) cells were plated in triplicate into 6-well plates for ~24 h in a humidified atmosphere of 5% CO₂ overnight at 37°C, followed by exposure to paclitaxel for 72 h at 37°C. Subsequently, cells were cultured for up to 3 weeks (BT-474) or 2 weeks (MDA-MB-231), during which time fresh medium was added every 3 days. For the colony formation assay, plates were fixed using 4% paraformaldehyde for 10 min at room temperature, stained with 0.1% crystal violet for 15 min at room temperature and counted under a light microscope (magnification, ×40; Nikon Corporation, Tokyo, Japan). Colonies with >50 cells were counted per well.

A total of 16 six-week-old BALB/c female nude mice were obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) and maintained under specific pathogen-free conditions (constant temperature, 24-26°C; humidity, 40-70%; 12/12 h light/dark cycle; free access to food and water). BT-474-shBeclin1#1 or BT-474-shControl cells (1×10⁷ per mouse) were injected subcutaneously into the right flank of the corresponding mice previously implanted with 17β-estradiol pellets. The tumor volumes were monitored twice weekly. When the tumor volumes reached ~100 mm³, the animals were divided randomly into four groups of 4 mice each. Mice were then intraperitoneally injected with either control vehicle or paclitaxel (10 mg/kg) twice a week for a total of 8 times (4 weeks). Tumor size was monitored using calipers and calculated using the formula: Length x width²/2. The rate of tumor inhibition was evaluated by the formula: (volume of the untreated group - the volume of the paclitaxel treatment group)/volume of the untreated group ×100%). All animal experiments were performed following the National Institutes of Health guide for the care and use of laboratory animals (36), and reviewed and ethically approved by the Ethics Committee of Navy General Hospital (Beijing, China). No significant animal weight loss was observed during the in vivo experiments. When the control tumors reached 1,000 mm³, all mice were sacrificed by cervical dislocation subsequent to being anesthetized with sodium pentobarbital intraperitoneally (60 mg/kg). Subsequently, the tumors were collected, fixed in 10% formalin for 48 h at room temperature and immunohistochemically analyzed.

Mouse tumor tissues were embedded in paraffin and sectioned into 5 µm slices. Serial sections were dewaxed in xylene and hydrated with graded concentrations of ethanol (100, 95, 90, 80 and 70%). Epitopes were retrieved using microwave in boiling 0.01 M sodium citrate buffer (pH 6.0) for 10 min at 98°C. Slides were treated with 3% H₂O₂ for 10 min, blocked with 5% goat serum for 1 h at room temperature and incubated with primary antibodies against Beclin1 (cat. no. sc-11427; 1:100 dilution; Santa Cruz Biotechnology, Inc.) and cleaved caspase-3 (cat. no. 9664; 1:200 dilution; Cell Signaling Technology, Inc.) at 4°C overnight. The signal was detected following incubation with a commercial immunoglobulin G horseradish peroxidase-conjugated secondary antibody polymer detection system (cat. no. PV-9001; OriGene Technologies, Inc., Beijing, China) for 20 min at 37°C and 3′3-diaminobenzidine substrate (OriGene Technologies, Inc.). Finally, the slides were counterstained with 0.5% hematoxylin (Beyotime Institute of Biotechnology, Shanghai, China) for 2 min at room temperature and observed by a light microscopy (magnification, ×200; Eclipse Ti-U; Nikon Corporation). Slides incubated without the primary antibody served as the negative controls.

Gene expression data (accession nos. GSE22513 and GSE25066) were retrieved from the National Centre for Biotechnology Information Gene Expression Omnibus (GEO) database (37). Beclin1 levels were analyzed in patients with pathologic complete response (pCR), which was defined as the lack of any invasive cancer, and non-pCR, which was defined as any viable tumor in breast or lymph nodes (partial response) or disease progression. pCR and non-pCR were determined from the primary pathologic slide selected at the time of surgery by a breast pathologist. Details of patients and microarray data were described previously (38,39). Expression data from GSE22513 were RMA summarized, quantile normalized and baseline transformed, making use of the median of all samples, using GeneSpring GX software (version 10.0.2; Agilent Technologies, Inc., Santa Clara, CA, USA). Data from GSE25066 were normalized using the MAS5 algorithm, mean centered to 600 and log² transformed.

SPSS version 19.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analyses. Statistical significance of the data was analyzed using a two-sample Student’s t-test (for two groups) or a one-way analysis of variance followed by Bonferroni’s post hoc test (for multiple groups). Data were presented as mean ± standard deviation (SD). The error bars represent SD values from three independent experiments unless otherwise indicated. P<0.05 was considered to indicate a statistically significant difference.

Initially, the effect of paclitaxel on BT-474 breast cancer cells was examined by exposing cells to various concentrations of the drug for 72 h. As the paclitaxel dose increased, the viability of BT-474 cells significantly decreased, demonstrating a dose-dependent drug effect (P<0.01; Fig. 1A). Next, a drug concentration of 0.5 µM was selected, which was known to result in an ~50% inhibition of cell viability, in order to analyze the time effect. As presented in Fig. 1B, paclitaxel significantly suppressed BT-474 cell proliferation in a time-dependent manner, as revealed by using a CCK-8 assay (P<0.001). In order to further determine whether paclitaxel-induced cytotoxicity was associated with apoptosis, an Annexin V-FITC apoptotic assay was performed, and programmed cell death was analyzed using the bright green FITC fluorescence detection of flow cytometry. The results revealed that 0.5 µM paclitaxel induced Annexin V-positive cells in 6.20±1.19, 10.54±1.32, 19.42±1.41 and 32.81±2.43% of total cells after 0, 24, 48 and 72 h of incubation (Fig. 1C and D), respectively, demonstrating that apoptosis was enhanced with time, with significant differences observed at the 48 and 72 h mark compared with untreated cells (P<0.001). Furthermore, apoptotic proteins were detected using western blot analysis. It was observed that the significantly increased cleavage of caspase-3 and PARP occurred 24 h after paclitaxel treatment in BT-474 cells compared with untreated cells for caspase-3 (P<0.05; Fig. 1E) and after 48 h for PARP (P<0.001; Fig. 1F). These results suggest that paclitaxel-induced apoptosis occurs through activation of caspases.

As the function of Beclin1 in paclitaxel-mediated cytotoxicity may be cancer type t dependent (25-31), the present study initially detected the expression of Beclin1 following paclitaxel treatment by western blot analysis. It was revealed that paclitaxel down-regulated Beclin1 protein levels in BT-474 cells in a dose- and time-dependent manner, with significant differences observed at the 0.5 µM dose (P<0.01) and 4 h of treatment (P<0.05) compared with the untreated cells (Fig. 2A and B). To assess the importance of Beclin1 alteration subsequent to paclitaxel treatment, the specific shRNA-mediated knockdown of the endogenous Beclin1 in BT-474 cells was performed. The effects of two different shRNAs on Beclin1 were evaluated using RT-qPCR and western blot analysis. The mRNA expression levels of Beclin1 in the shBeclin1 groups were demonstrated to be significantly lower in comparison with that in the empty vector control group (P<0.001; Fig. 3A). These results indicated that the transfection efficiency of shBeclin1 was high. Furthermore, the two shRNAs were able to significantly decrease the mRNA and protein expression levels of Beclin1, compared with the shControl group (P<0.001; Fig. 3A and B). Next, the present study assessed the responses of BT-474 cells to treatments with different concentrations of paclitaxel for 72 h using a CCK-8 assay. As presented in Fig. 3C, the curative effect of paclitaxel (at a concentration of 0.1 µM or above) was significantly enhanced due to Beclin1 knockdown, compared with the control group that was infected with shControl in BT-474 cells (P<0.05). Furthermore, the time-dependent effect of BT-474 cells responses to paclitaxel was analyzed using a fixed drug concentration of 0.5 µM. Similar to concentration effects, the inhibition of Beclin1 resulted in the significant potentiation of cytotoxicity of paclitaxel in a time-dependent manner compared with the control group (P<0.01; Fig. 3D). To exclude effects associated with cell specificity, an additional breast cancer cell line, MDA-MB-231, was used to demonstrate the effects of Beclin1 silencing on paclitaxel response. As presented in Fig. 4, suppression of Beclin1 also enhanced paclitaxel-mediated cytotoxicity in MDA-MB-231 cells compared with the control cells. In accordance with the results from the CCK-8 assay, Beclin1 knockdown significantly enhanced the inhibition of colony formation by paclitaxel compared with the shControl group in BT-474 and MDA-MB-231 cells (P<0.01; Fig. 5). In summary, the results of the present study demonstrated that Beclin1 protected breast cancer cells from the cytotoxicity effect of paclitaxel.

As the present results and previous research has demonstrated that paclitaxel is able to induce apoptosis in breast cancer cells (40-42), the present study further aimed to identify whether Beclin1 knockdown (which resulted in the enhancement of the effect of paclitaxel) is associated with increased apoptosis. Analysis of apoptosis, performed using flow cytometry, discovered that Beclin1 knockdown alone was not able to significantly enhance apoptosis, compared with the shControl group in BT-474 and MDA-MB-231 cells (Fig. 6). However, the suppression of Beclin1 significantly enhanced paclitaxel-induced apoptosis in BT-474 and MDA-MB-231 cells compared with the shControl group (P<0.05; Fig. 6). Furthermore, the present study determined the expression of apoptosis-associated proteins by western blot analysis. In accordance with the aforementioned results, the cleavage of caspase-3 (P<0.05) and PARP (P<0.01) were significantly elevated in the Beclin1 knockdown plus paclitaxel treatment groups, compared with the paclitaxel-treated alone control group in breast cancer cells (Fig. 7). Collectively, these data suggested that Beclin1 inhibited caspase-dependent apoptosis induced by paclitaxel in breast cancer cells.

The results obtained above demonstrate that Beclin1 knockdown enhanced paclitaxel-mediated cytotoxicity by inducing apoptosis in vitro. The present study further aimed to reveal whether the therapeutic efficacy of paclitaxel may be improved following Beclin1 suppression in vivo. As presented in Fig. 8A, tumor growth in the Beclin1-knockdown group was significantly inhibited compared with the shControl group (P<0.05), subsequent to paclitaxel treatment in the BT-474 xenograft model. The rate of tumor inhibition of paclitaxel was also calculated. It was established that Beclin1 knockdown significantly improved the antitumor efficacy of paclitaxel, compared with the shControl (P<0.05; Fig. 8B). Tumor specimens collected at the end of the experiments were analyzed using IHC staining (Fig. 8C). Suppression of Beclin1 in the shBeclin1#1 group was confirmed in vivo (Fig. 8C, upper row). Furthermore, IHC staining revealed that the cleaved caspase-3 positive cells in the Beclin1 knockdown tumor types were elevated compared with control tumor types in response to paclitaxel treatment (Fig. 8C, lower row). These results are consistent with the in vitro data. Altogether, these observations supported the hypothesis of Beclin1 serving a critical function in modulating the antitumor efficacy of paclitaxel in breast cancer cells in vivo.

To clarify whether levels of Beclin1 are associated with clinical response to paclitaxel therapy, the present study analyzed one GEO dataset (GSE22513) containing patients receiving neoadjuvant paclitaxel/radiation treatment (35). The results revealed that Beclin1 levels in non-pCR patients were significantly elevated, compared with pCR patients (P<0.001; Fig. 9A). Meanwhile, another GEO dataset (GSE25066), including patients treated with neoadjuvant chemotherapy containing paclitaxel, was analyzed (36). Compared to pCR patients, the expression levels of Beclin1 were significantly increased in non-pCR patients (P<0.001; Fig. 9B). These results suggest that Beclin1 may be a potential clinical predictor of the response of a patient to paclitaxel.